List your company

Software Engineering Issue Resolving on SWE-bench Verified

66Pass@1Context Weaver
33.44841.89950.3558.801Apr 15, 2026Apr 16, 2026Apr 18, 2026Apr 19, 2026Apr 21, 2026Apr 22, 2026Apr 24, 2026

Evaluation results

MethodLinks
Context WeaverModel=Claude Sonnet 4, Setting=Unified2026.0466
LLM SummarizationModel=Claude Sonnet 4, Setting=Unified2026.0464.2
MTL (I)Model Backbone=GPT-5-mini2026.0463.3
Sliding WindowModel=Claude Sonnet 4, Setting=Unified, window size=52026.0463.2
ZeroShotModel Backbone=GPT-5-mini2026.0462.3
MTL (W)Model Backbone=GPT-5-mini2026.0462
MTL (S)Model Backbone=GPT-5-mini2026.0461.3
MTL (T)Model Backbone=GPT-5-mini2026.0461
Sliding WindowModel=Gemini 3 Flash, Setting=Unified, window size=52026.0460.4
Context WeaverModel=GPT-5, Setting=Hybrid2026.0458.6
Context WeaverModel=Gemini 3 Flash, Setting=Unified2026.0458.4
Sliding WindowModel=GPT-5, Setting=Unified, window size=52026.0457.4
Context WeaverModel=GPT-5, Setting=Unified2026.0456.8
LLM SummarizationModel=Gemini 3 Flash, Setting=Unified2026.0456.8
LLM SummarizationModel=GPT-5, Setting=Unified2026.0446.7
MTL (I)Model Backbone=DeepSeek V3.22026.0446.3
ZeroShotModel Backbone=DeepSeek V3.22026.0442.3
MTL (I)Model Backbone=Qwen3-Coder-480B-A35B-Instruct2026.0437
ZeroShotModel Backbone=Qwen3-Coder-480B-A35B-Instruct2026.0434.7